NMNeda Maleki

Neda Maleki

*This is an algorithmically generated author record based on publicly available information.

Neda Maleki has a total of 15 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Online Course Improvement Through GPT-4: Monitoring Student Engagement and Dynamic FAQ Generation

    by Oxana Lundström, Neda Maleki, Fredrik Ahlgren

    Artificial Intelligence (AI), specifically in language processing, is increasingly recognized as an invaluable educational tool. The Large Language Model (LLM) GPT, developed by OpenAI, is an advanced machine learning tool that utilizes deep learning for human-like text comprehension and generation. This study uses OpenAI's GPT-4 to enhance an online Internet of Things (loT) course at Linnaeus University. We analyzed 12,000+ messages on an online communication platform spanning four years. We compare traditional Natural Language Processing (NLP) techniques to Generative AI for understanding student feedback and issues, inspiring project ideas, and promoting student engagement. We provide a combined approach to monitor the sentiment or mood of the students' communications over the timeline of the course. Moreover, we show how to use LLM to refine the FAQ generation and decipher student feedback for course refinement. We demonstrate how to generate optimal prompts and prepare the data to apply LLMs effectively. Our research reinforces that strategic use of LLMs, like GPT-4, can revolutionize remote learning by lessening lecturer workload and boosting student satisfaction and engagement. Our future work aims to further leverage AI models across remote engineering education. One potential direction is developing an AI-powered bot for online platforms to facilitate real-time interaction, manage queries, encourage engagement, maintain FAQs, and enhance course outcomes.


  • Open Access

    Internet of Things for Wetland Conservation using Helium Network: Experience and Analysis

    by Arslan Musaddiq, Neda Maleki, Francis Palma, David Mozart, Tobias Olsson, Mustafa Omareen, Fredrik Ahlgren

    The Internet of Things (IoT), as a new paradigm of connected things or objects to the Internet, allows us to monitor the environment by collecting data in a wide spatial and temporal window. Especially the utilization of IoT has increased significantly since the development of the Long Range Wide Area Network (LoRaWAN). However, deploying LoRa gateways, maintaining network infrastructure, operational cost, and quality of service are challenging. Helium has emerged as one of the largest networks in terms of coverage for IoT devices to solve such problems. Helium is decentralized, cryptocurrency incentives-based network infrastructure replacing traditional service providers. However, due to network incentives, currently, it contains more hotspots compared to active users. This paper presents our experience and analysis of deploying IoT devices for real-world applications using the Helium network. We present experiences from the IoT device’s deployment for wetland conservation in southern Sweden.


  • Open Access

    TMaR: a two-stage MapReduce scheduler for heterogeneous environments

    by Neda Maleki, Hamid Reza Faragardi, Amir Masoud Rahmani, Mauro Conti, Jay Lofstead

    In the context of MapReduce task scheduling, many algorithms mainly focus on the scheduling of Reduce tasks with the assumption that scheduling of Map tasks is already done. However, in the cloud deployments of MapReduce, the input data is located on remote storage which indicates the importance of the scheduling of Map tasks as well. In this paper, we propose a two-stage Map and Reduce task scheduler for heterogeneous environments, called TMaR. TMaR schedules Map and Reduce tasks on the servers that minimize the task finish time in each stage, respectively. We employ a dynamic partition binder for Reduce tasks in the Reduce stage to lighten the shuffling traffic. Indeed, TMaR minimizes the makespan of a batch of tasks in heterogeneous environments while considering the network traffic. The simulation results demonstrate that TMaR outperforms Hadoop-stock and Hadoop-A in terms of makespan and network traffic and achieves by an average of 29%, 36%, and 14% performance using Wordcount, Sort, and Grep benchmarks. Besides, the power reduction of TMaR is up to 12%.


  • SoFA: A Spark-oriented Fog Architecture

    by Neda Maleki, Mohammad Loni, Masoud Daneshtalab, Mauro Conti, Hossein Fotouhi

    Fog computing offers a wide range of service levels including low bandwidth usage, low response time, support of heterogeneous applications, and high energy efficiency. Therefore, real-time embedded applications could potentially benefit from Fog infrastructure. However, providing high system utilization is an important challenge of Fog computing especially for processing embedded applications. In addition, although Fog computing extends cloud computing by providing more energy efficiency, it still suffers from remarkable energy consumption, which is a limitation for embedded systems. To overcome the above limitations, in this paper, we propose SoFA, a Spark-oriented Fog architecture that leverages Spark functionalities to provide higher system utilization, energy efficiency and scalability. Compared to the common Fog computing platforms where edge devices are only responsible for processing data received from their IoT nodes, SoFA leverages the remaining processing capacity of all other edge devices. To attain this purpose, SoFA provides a distributed processing paradigm by the help of Spark to utilize the whole processing capacity of all the available edge devices leading to increase energy efficiency and system utilization. In other words, SoFA proposes a near-sensor processing solution in which the edge devices act as the Fog nodes. In addition, SoFA provides scalability by taking advantage of Spark functionalities. According to the experimental results, SoFA is a power-efficient and scalable solution desirable for embedded platforms by providing up to 3.1x energy efficiency for the Word-Count benchmark compared to the common Fog processing platform.